Field manual · AI for authorized security testing · Edition November 2026 · Updated 2026-11-18

Tool · Coding agents as copilots

Claude Code, Codex and OpenCode for an authorized pentest

Claude Code, Codex and OpenCode become a pentest copilot when you point them at your model and put the written scope in the project.

Claude Code reads ANTHROPIC_BASE_URL, Codex reads a provider block, and OpenCode reads opencode.json.

Updated November 18, 2026Authorized work only
01Fit

A copilot for the operator, not a scanner

Use a coding agent when a person still chooses the next check. Use Strix or Shannon for a whole application test. Put Kimi K3.1 Pro in the default slot and write the hosts in SCOPE.md.

02Claude Code

Claude Code reads ANTHROPIC_BASE_URL

The settings file maps every built-in alias onto a model id.

The base URL is the host, with no /v1, because the client adds /v1/messages itself. Claude Code guide.

~/.claude/settings.jsonsettings.json
{
  "apiKeyHelper": "echo $SILK_API_KEY",
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.silkcompute.com",
    "ANTHROPIC_DEFAULT_MODEL": "kimi-k3.1-pro-uncensored",
    "ANTHROPIC_DEFAULT_FABLE_MODEL": "kimi-k3.1-pro-uncensored",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "kimi-k3.1-pro-uncensored",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "kimi-k3.1-pro-uncensored",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "deepseek-v4.1-flash-1017-uncensored",
    "CLAUDE_CODE_MAX_CONTEXT_TOKENS": "1048576"
  },
  "modelPicker": {
    "replaceBuiltInOptions": true,
    "options": [
      { "model": "kimi-k3.1-pro-uncensored", "label": "Kimi K3.1 Pro" },
      {
        "model": "deepseek-v4.1-flash-1017-uncensored",
        "label": "DeepSeek V4.1 Flash 1017"
      },
      {
        "model": "deepseek-v4.1-flash-1017-fast-uncensored",
        "label": "DeepSeek V4.1 Flash 1017 Fast"
      },
      { "model": "glm-5.3-uncensored", "label": "GLM-5.3" },
      { "model": "glm-5.3-flash-uncensored", "label": "GLM-5.3 Flash" },
      { "model": "hy4-uncensored", "label": "Hy4" }
    ]
  }
}

Get a key. The primary model is kimi-k3.1-pro-uncensored. The Haiku alias is the lowest output price, and Opus and Fable point at the highest.

03Codex

Codex reads a provider block

Codex reads the model list from the API. web_search is off. Codex guide.

~/.codex/config.tomlconfig.toml
model = "kimi-k3.1-pro-uncensored"
model_provider = "silkcompute"
web_search = "disabled"

[model_providers.silkcompute]
name = "Silk Compute"
base_url = "https://api.silkcompute.com"
env_key = "SILK_API_KEY"
model_catalog_url = "https://api.silkcompute.com/codex/models"
04OpenCode

OpenCode reads an OpenAI-compatible provider

The provider block names @ai-sdk/openai-compatible, the base URL without /v1, and one entry per model. OpenCode guide. Cursor uses the same idea in its own file. Cursor guide.

opencode.jsonopencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "model": "silkcompute/kimi-k3.1-pro-uncensored",
  "provider": {
    "silkcompute": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Silk Compute",
      "options": {
        "baseURL": "https://api.silkcompute.com",
        "apiKey": "{env:SILK_API_KEY}"
      },
      "models": {
        "kimi-k3.1-pro-uncensored": {
          "name": "Kimi K3.1 Pro",
          "tool_call": true,
          "interleaved": { "field": "reasoning_content" },
          "modalities": { "input": ["text", "image", "video"], "output": ["text"] },
          "limit": { "context": 1048576, "output": 131072 },
          "cost": { "input": 3, "output": 8, "cache_read": 0.3 },
          "variants": {
            "none": { "reasoningEffort": "none" },
            "low": { "reasoningEffort": "low" },
            "high": { "reasoningEffort": "high" },
            "max": { "reasoningEffort": "max" }
          }
        },
        "deepseek-v4.1-flash-1017-uncensored": {
          "name": "DeepSeek V4.1 Flash 1017",
          "tool_call": true,
          "interleaved": { "field": "reasoning_content" },
          "modalities": { "input": ["text", "image"], "output": ["text"] },
          "limit": { "context": 1048576, "output": 131072 },
          "cost": { "input": 0.24, "output": 0.9, "cache_read": 0.038 },
          "variants": {
            "none": { "reasoningEffort": "none" },
            "low": { "reasoningEffort": "low" },
            "high": { "reasoningEffort": "high" },
            "max": { "reasoningEffort": "max" }
          }
        },
        "deepseek-v4.1-flash-1017-fast-uncensored": {
          "name": "DeepSeek V4.1 Flash 1017 Fast",
          "tool_call": true,
          "interleaved": { "field": "reasoning_content" },
          "modalities": { "input": ["text", "image"], "output": ["text"] },
          "limit": { "context": 1048576, "output": 131072 },
          "cost": { "input": 0.24, "output": 0.9, "cache_read": 0.05 },
          "variants": {
            "none": { "reasoningEffort": "none" },
            "low": { "reasoningEffort": "low" },
            "high": { "reasoningEffort": "high" },
            "max": { "reasoningEffort": "max" }
          }
        },
        "glm-5.3-uncensored": {
          "name": "GLM-5.3",
          "tool_call": true,
          "interleaved": { "field": "reasoning_content" },
          "modalities": { "input": ["text"], "output": ["text"] },
          "limit": { "context": 1048576, "output": 131072 },
          "cost": { "input": 2, "output": 4, "cache_read": 0.3 },
          "variants": {
            "none": { "reasoningEffort": "none" },
            "low": { "reasoningEffort": "low" },
            "high": { "reasoningEffort": "high" },
            "max": { "reasoningEffort": "max" }
          }
        },
        "glm-5.3-flash-uncensored": {
          "name": "GLM-5.3 Flash",
          "tool_call": true,
          "interleaved": { "field": "reasoning_content" },
          "modalities": { "input": ["text", "image"], "output": ["text"] },
          "limit": { "context": 1048576, "output": 131072 },
          "cost": { "input": 0.24, "output": 1, "cache_read": 0.06 },
          "variants": {
            "none": { "reasoningEffort": "none" },
            "low": { "reasoningEffort": "low" },
            "high": { "reasoningEffort": "high" },
            "max": { "reasoningEffort": "max" }
          }
        },
        "hy4-uncensored": {
          "name": "Hy4",
          "tool_call": true,
          "interleaved": { "field": "reasoning_content" },
          "modalities": { "input": ["text", "image"], "output": ["text"] },
          "limit": { "context": 1048576, "output": 65536 },
          "cost": { "input": 1.6, "output": 3.2, "cache_read": 0.3 },
          "variants": {
            "none": { "reasoningEffort": "none" },
            "low": { "reasoningEffort": "low" },
            "high": { "reasoningEffort": "high" },
            "max": { "reasoningEffort": "max" }
          }
        }
      }
    }
  }
}
Rules of engagement

Only on systems you own, a signed scope, a CTF or a lab. A custom model does not widen the scope.

05Prompt

Ask for the plan, and point at the scope file

Prompt

Read SCOPE.md. Draft the test plan for the in-scope hosts only. Name what to review and what evidence to collect. Do not write exploit code. Skip any step outside SCOPE.md.